TWI479042B - A method of plasma vapour deposition - Google Patents

Abstract

This invention consists of a method of plasma vapour deposition of metal into a recess in a workpiece. The method achieves re-sputtering of the metal at the base of the recess with a sputter gas by utilising a mixture of Ar and He and/or Ne as the sputter gas with a ratio of He and/or Ne:Ar of at least about 10:1.

Description

Plasma vapor deposition method Field of invention

SUMMARY OF THE INVENTION The present invention is directed to a method of vapor depositing a metal plasma into a recess in a working component in a manner that accomplishes resputtering the metal of a recessed substrate onto the sidewall.

Background of the invention

It is known to use a ionized metal sputtering technique to include a high power unbalanced magnetron discharge source. The metal ions are attracted to a substrate formed in a recess in a working component (e.g., a semiconductor wafer) using a DC bias caused by application of RF power to the wafer carrier. This imparts improved bottom and sidewall coverage in the recess. Further improvements are known by re-sputtering the metal that has been deposited on the substrate to the bottom and intermediate portions of the sidewall of the recess. In order to complete re-sputtering, a DC bias in the range of 50 to 500 volts is required. The higher the DC bias voltage, the better the sputtering and the improved sidewall coverage.

Even so, the results today are still not enough for all purposes.

Summary of invention

From one point of view, the present invention comprises a method of vapor-depositing a metal plasma into a recess in a working member in such a manner as to re-sputter the metal of a recessed substrate onto the sidewall. A metal target is sputtered with a sputtering gas characterized by a mixture of argon and helium in the sputtering gas system when the ratio of helium to argon is at least about 10:1.

Preferably the ratio is about 20:1.

The flow rate of argon is <10 sccm and the flow rate of helium is >100 sccm. Thus, for example, the helium flow rate is about 140 sccm and the argon flow rate is about 7 sccm.

In any of the above examples, the metal is copper.

In another embodiment, the method is characterized by a mixing system of helium and argon such that the target current density is at least about 0.035 A/cm ² and preferably about 0.037 A/cm ² .

Although the invention has been defined above, it will be appreciated that it includes any inventive combination of the features set forth above or described below.

Simple illustration

The invention may be embodied in a variety of ways, and the specific embodiments will now be described by way of example with reference to the accompanying drawings in which: FIG. 1 is a cross-sectional view of a DC magnetron ionization deposition sputtering system; Figure 2 is a table showing secondary electron emission as a function of the bombardment energy and gas type of ions incident on the molybdenum and tungsten targets. This pattern is taken from Glow Discharge Processes, Chapmen, Wiley and Sons 1980; Figure 3 illustrates the change in flow rate versus target current and voltage during the addition of helium to the argon sputtering process; Figure 4 shows A standard argon method covers the sidewalls of a 5:1 trench structure; Figure 5 shows an equivalent view of an optimized helium/argon method and Figure 4; and Figure 6 shows a reduction of flaws. An equivalent view of the helium/argon method; Figures 6(a), (b), and (c) show the effect of the helium/argon flow on the deposition rate (correction factor), pressure, and in-wafer non-uniformity, respectively; Figure 7 is an equivalent view of another hernia formula and Figure 5.

Detailed description of the preferred embodiment

In an unbalanced DC magnetron ionization sputter deposition system, metal and gas ionization is achieved by metal and gas atoms colliding with secondary electrons generated on the surface of the target. The secondary electrons are accelerated and reach high energy by a negative bias provided to the surface of the target, typically between 200 and 1000 eV.

These high energy secondary electrons collide with metals and gas atoms or molecules and generate ions. Thus, if the amount of secondary electrons emitted by the target can be increased, there will be an increase in the ionized portion, i.e., the ratio of ions to neutral species of any kind.

As can be seen in Figure 2, the secondary electronic properties for different sputtering gases vary depending on the sputtering gas used. For example, tantalum and niobium have a higher level of secondary electron emission than gases such as argon and helium. Applicants have learned that a sputtering system that replaces argon with helium or neon may result in more secondary electrons emitted by the target and the resulting ionized portion.

This effect has been investigated by the Applicant by using a copper target to progressively introduce increased levels of turbulence and reduce argon flow, and it can be seen that as the enthalpy increases and argon decreases, due to the surface of the target Sub-electron formation causes the current to increase. Low levels of argon present have been found to be necessary to maintain film density quality.

experiment

The applicant then proceeded to an experiment using the apparatus of Figure 1 and the method conditions shown in Table 1 below.

This compares the standard argon-only method operating on the apparatus as in Figure 1 and a helium/argon process in which the argon flow is 7 sccm and the turbulent flow is 140 sccm. The overlay resulting from this first method is shown in Figure 4, however for this second method, the overlay is illustrated in Figure 5. The resulting step coverage is summarized in Table 2 below.

It is seen that the helium/argon method shows an 5% absolute, 30% proportional increase in the value of the sidewall covering system beyond the argon only method.

As can be seen from Table 1, for the helium/argon method, the platform DC bias is only increased by 30V (12%) compared to the argon only method (standard), although the RF power applied to the platform has increased. 165W (50%). This shows that the helium/argon process has a higher level of ionization because the positive ions in the plasma will tend to reduce the negative DC bias achieved at the platform.

In Figure 6, the same method was performed, but this time with a turbulent flow of 75 sccm and the results are summarized in Table 3 below.

It can be seen that comparing the 140 氦 / 7 argon method and the argon only method, reducing turbulence causes a reduction in sidewall coverage. This is because in this experiment, the secondary electrons generated in the plasma were reduced by lower turbulence.

Experiments have shown that <10 Sccm of argon is required to maintain the plasma and to cause sputtering of the copper target to occur. Conversely, a turbulence system of >100 Sccm is required to assist in maintaining the plasma and to provide additional secondary electrons to enhance the sputtering effect at the bottom of the structure and thus improve sidewall coverage.

In further experiments, the deposition rate, stress, and in-wafer non-uniformity were monitored as the flow of change, and the results were shown in Section 6(a), respectively. To 6 (c). It can be seen that although the deposition rate decreases with the addition of ruthenium, the 7 argon/140 方法 method is still sufficiently higher than the production purpose. This combination has particularly good stress values and predictable uniformity. It will be appreciated that reduced film stress helps to prevent the copper film from peeling off from the underlying material.

Figure 7 illustrates a reduced turbulent flow with an increased argon flow and the step coverage is again reduced, as presented in Table 4 below.

Thus, in summary, it can be seen that considerable improvement can be achieved by having a turbulent flow rate above about 100 sccm and an argon flow rate below about 10 sccm. For the reasons indicated above, the 7 argon/140 Torr method results in a particularly preferred method.

In fact, it may be more generally stated that, in terms of the results achieved, the partial pressures of these gases should be relatively constant at the corresponding constants, however the actual flow rate may be in the chamber and chamber. There are changes. Table 5 below presents the equivalent partial pressures for these experimental flow rates.

Metal-based copper used in these experiments. The process of the invention should be similar for titanium, tantalum, gold or rhenium.

Figure 1 is a cross-sectional view of a DC magnetron ionization deposition sputtering system; Figure 2 is a diagram showing secondary electron emission as a function of bombardment energy and gas type for ions incident on molybdenum and tungsten targets. Form. This pattern is taken from Glow Discharge Processes, Chapmen, Wiley and Sons 1980; Figure 3 illustrates the change in flow rate versus target current and voltage during the addition of helium to the argon sputtering process; Figure 4 shows A standard argon method is applied to the sidewall of a 5:1 trench structure; Figure 5 shows an equivalent view of an optimized helium/argon method and Figure 4; Figure 6 shows an equivalent view of a helium/argon method with reduced helium; 6(a), (b) (c) shows the effect of the helium/argon flow on the deposition rate (correction factor), stress, and in-wafer non-uniformity, respectively; and Figure 7 is the equivalent of another helium formulation and Figure 5. View.

Claims (6)

A method for vapor-depositing a metal plasma into a recess in a working part by sputtering a metal target with a sputtering gas to achieve re-sputtering the metal of the substrate at the recess to the sidewall Characterized by the sputter gas system being a mixture of argon (Ar) and helium (He) and/or neon (Ne) wherein the ratio of helium and/or neon:argon is at least about 10:1. The method of claim 1, wherein the ratio is about 20:1. The method of claim 1, wherein the flow rate of the argon is <10 sccm and the flow rate of the helium and/or helium is >100 sccm. The method of claim 3, wherein the turbulence rate is about 140 sccm and the argon flow rate is about 7 sccm. The method of any one of claims 1 to 4, wherein the metal is selected from the group consisting of copper, titanium, tantalum, gold or ruthenium. The method of claim 1, wherein the mixture of ruthenium and/or krypton and argon is such that the target current density is 0.035 A/cm ² .